From 767103e410d29c63f7652dc9246d45c7996af8ce Mon Sep 17 00:00:00 2001
From: Loi Nguyen <1948922+lntutor@users.noreply.github.com>
Date: Mon, 20 Jul 2026 03:20:40 +0700
Subject: [PATCH] fix(html): reduce inline images to alt text under markdownify
 1.x

markdownify >= 1.0 changed the convert_* calling convention: it passes
parent_tags (containing the pseudo-tag '_inline' for headings/table
cells) instead of the old positional convert_as_inline boolean.
_CustomMarkdownify.convert_img still declared convert_as_inline and
swallowed parent_tags into **kwargs, so convert_as_inline was always
False and the alt-text reduction never fired. An image inside a heading
or table cell therefore leaked full image markdown -- e.g.
'<h1>Hello <img src=a.png alt=World></h1>' became '# Hello ![World](a.png)'
instead of '# Hello World', and a data: URI in a table cell could break
the row.

Switch the override to check '_inline' in parent_tags, matching base
markdownify. Block (paragraph) images are unaffected.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01N6RtoHuxrDqTUo9Mw9h4Cv
---
 .../src/markitdown/converters/_markdownify.py |  9 +++++--
 packages/markitdown/tests/test_module_misc.py | 26 +++++++++++++++++++
 2 files changed, 33 insertions(+), 2 deletions(-)

diff --git a/packages/markitdown/src/markitdown/converters/_markdownify.py b/packages/markitdown/src/markitdown/converters/_markdownify.py
index 19e8a2984..ca1358fab 100644
--- a/packages/markitdown/src/markitdown/converters/_markdownify.py
+++ b/packages/markitdown/src/markitdown/converters/_markdownify.py
@@ -86,11 +86,16 @@ def convert_img(
         self,
         el: Any,
         text: str,
-        convert_as_inline: Optional[bool] = False,
+        parent_tags: Optional[Any] = None,
         **kwargs,
     ) -> str:
         """Same as usual converter, but removes data URIs"""
 
+        # markdownify >= 1.0 passes `parent_tags` (containing the pseudo-tag
+        # "_inline" for headings/table cells) instead of the old
+        # `convert_as_inline` boolean; check that so inline images still reduce
+        # to their alt text.
+        parent_tags = parent_tags or set()
         alt = el.attrs.get("alt", None) or ""
         src = el.attrs.get("src", None) or el.attrs.get("data-src", None) or ""
         title = el.attrs.get("title", None) or ""
@@ -98,7 +103,7 @@ def convert_img(
         # Remove all line breaks from alt
         alt = alt.replace("\n", " ")
         if (
-            convert_as_inline
+            "_inline" in parent_tags
             and el.parent.name not in self.options["keep_inline_images_in"]
         ):
             return alt
diff --git a/packages/markitdown/tests/test_module_misc.py b/packages/markitdown/tests/test_module_misc.py
index 4d62e4919..854fc8450 100644
--- a/packages/markitdown/tests/test_module_misc.py
+++ b/packages/markitdown/tests/test_module_misc.py
@@ -552,3 +552,29 @@ def test_markitdown_llm() -> None:
         test()
         print("OK")
     print("All tests passed!")
+
+
+def test_inline_image_in_heading_reduced_to_alt() -> None:
+    from markitdown.converters._html_converter import HtmlConverter
+
+    md = HtmlConverter().convert_string('<h1>Hello <img src="a.png" alt="World"></h1>').markdown
+    assert md.strip() == "# Hello World", repr(md)
+
+
+def test_inline_image_in_table_cell_reduced_to_alt() -> None:
+    from markitdown.converters._html_converter import HtmlConverter
+
+    html = (
+        "<table><tr><th>Pic</th><th>Name</th></tr>"
+        '<tr><td><img src="a.png" alt="AA"></td><td>bob</td></tr></table>'
+    )
+    md = HtmlConverter().convert_string(html).markdown
+    assert "| AA | bob |" in md, repr(md)
+    assert "![AA]" not in md, repr(md)
+
+
+def test_block_image_still_rendered() -> None:
+    from markitdown.converters._html_converter import HtmlConverter
+
+    md = HtmlConverter().convert_string('<p><img src="a.png" alt="x"></p>').markdown
+    assert "![x](a.png)" in md, repr(md)